JP6323511B2 - Driver condition detection apparatus and method - Google Patents

Description

  The technology disclosed herein relates to a driver physical condition detection apparatus and method for detecting a physical condition of a driver.

  One of the causes of traffic accident deaths is a sudden change in the physical condition of the driver while driving. Factors that cause a sudden change in the physical condition of the driver include various diseases such as a cerebrovascular disease and a heart disease, and the state of the driver who cannot continue driving due to the sudden change in the physical condition is not constant. Conventionally, a technique for detecting a sudden change in the physical condition of a driver is known (for example, see Patent Document 1). In the technique described in Patent Document 1, an abnormality of the driver is estimated based on the driving posture of the driver, and an attempt is made to detect a sign of deterioration in physical condition.

Japanese Patent Laying-Open No. 2015-021912

  In general, a large change in the driving posture of the driver often appears after an abnormal state has progressed to some extent. However, in order to ensure the safety of the driver, it is necessary to detect an abnormality in the physical condition of the driver at an early stage before the abnormal state proceeds.

  The technique disclosed here aims to detect an abnormality in the physical condition of the driver at an early stage before the abnormal state progresses.

One aspect of the technology disclosed herein is a driver physical condition detection device that detects a physical condition of a driver who drives a vehicle, the vehicle detection unit detecting a change in movement of the vehicle, and the driving A driver detection unit that detects a change in the driver's movement, a calculation unit that calculates a tracking degree of the change in the driver's movement with respect to the change in the driving motion, and a physical condition of the driver based on the tracking degree A physical condition determination unit that executes a determination process for determining whether or not the vehicle is abnormal, and the calculation unit calculates, as the follow-up degree, a time delay of the change in the driver's movement with respect to the change in the movement during the travel. calculated, the physical condition judging unit, the when the time delay is less than the reference time, the physical condition of the driver is shall be determined to be abnormal.

In this aspect, the calculation unit calculates the degree of follow-up of the change in the driver's movement with respect to the change in the movement while the vehicle is running. Based on the degree of follow-up, the physical condition determining unit determines whether or not the driver's physical condition is abnormal. The degree of follow-up of the change in the driver's movement relative to the change in movement while the vehicle is running is different from when the physical condition is normal because the muscle strength of the neck or upper body decreases when the driver's physical condition begins to become abnormal It will be a thing. Therefore, according to this aspect, it is possible to detect an abnormality in the physical condition of the driver at an early stage before the abnormal state proceeds.
Further, in this aspect, the time delay of the change in the driver's movement relative to the change in the movement while the vehicle is traveling is calculated by the calculation unit as the tracking degree. When the time delay is equal to or less than the reference time, the physical condition determination unit determines that the physical condition of the driver is abnormal. When the muscle strength of the neck or upper body decreases at the stage where the driver's physical condition begins to become abnormal, the time delay of the change in the driver's movement with respect to the change in the movement while the vehicle is traveling is reduced. Therefore, according to this aspect, it is possible to detect an abnormality in the physical condition of the driver at an early stage before the abnormal state proceeds.

  In the above aspect, for example, the vehicle detection unit may detect a change in movement during the traveling in the left-right direction of the vehicle. The driver detection unit may detect a change in the movement of the driver in the left-right direction. The calculation unit may calculate a follow-up degree of a change in the driver's movement in the left-right direction with respect to a change in the movement in the left-right direction.

  In this aspect, the degree of follow-up of the change in the movement of the driver in the left-right direction of the vehicle with respect to the change in the movement in the left-right direction of the vehicle is calculated. The degree of follow-up of the change in the movement of the driver in the left-right direction of the vehicle relative to the change in the movement in the left-right direction of the vehicle is higher than that in the front-rear direction of the vehicle. Is significantly different from when normal. Therefore, according to this aspect, it is possible to detect an abnormality in the physical condition of the driver at an early stage before the abnormal state proceeds.

  In the above aspect, for example, the driver detection unit may detect a change in the center of gravity position of the driver as a change in the driver's movement. The calculation unit may calculate a follow-up degree of the change in the center of gravity position with respect to the change in movement during the traveling.

  In this aspect, a change in the driver's center of gravity is detected by the driver detection unit as a change in the driver's movement. The degree of follow-up of the change in the center of gravity position of the driver with respect to the change in the movement of the vehicle while traveling is calculated by the calculation unit. The degree of follow-up of the change in the center of gravity of the driver with respect to the change in the movement of the vehicle while the vehicle is running is because the muscle strength of the neck or upper body decreases when the driver's physical condition begins to become abnormal. It will be different. Therefore, according to this aspect, it is possible to detect an abnormality in the physical condition of the driver at an early stage before the abnormal state proceeds.

  In the above aspect, for example, the driver detection unit may detect acceleration of the driver's head as a change in the driver's movement. The calculation unit may calculate a tracking degree of acceleration of the head with respect to a change in movement during the traveling.

  In this aspect, the driver's head acceleration is detected by the driver detection unit as a change in the driver's movement. The degree of follow-up of the acceleration of the driver's head with respect to a change in movement while the vehicle is running is calculated by the calculation unit. The degree of follow-up of acceleration of the driver's head with respect to changes in the movement of the vehicle while the vehicle is running is because the muscle strength of the neck or upper body decreases when the driver's physical condition begins to become abnormal, and when the physical condition is normal It will be different. Therefore, according to this aspect, it is possible to detect an abnormality in the physical condition of the driver at an early stage before the abnormal state proceeds.

  In the above aspect, for example, the calculation unit may calculate the time delay by calculating a cross-correlation between the time change of the movement while traveling and the time change of the driver's movement.

  In this aspect, the cross-correlation between the time change of the movement of the vehicle while the vehicle is moving and the time change of the driver's movement is calculated by the calculation unit, and the time delay is calculated. Therefore, according to this aspect, the time delay can be calculated with high accuracy.

  In the above aspect, for example, the physical condition determination unit may execute the determination process only when the magnitude of the change in movement during the traveling is equal to or greater than a predetermined threshold.

  In this aspect, the physical condition determination unit executes the determination process only when the magnitude of the change in movement during the traveling of the vehicle is equal to or greater than a predetermined threshold. If the magnitude of the change in movement of the vehicle is less than a predetermined threshold, the driver's physical condition is normal with respect to the degree of follow-up of the change in movement of the driver with respect to the change in movement of the vehicle. It may not show a significant difference between certain cases and abnormal cases. Therefore, according to this aspect, it is possible to accurately determine whether or not the driver's physical condition is abnormal.

  In the above aspect, for example, the vehicle detection unit may include an acceleration sensor that detects acceleration of the vehicle as a change in movement during the traveling.

  In this aspect, the acceleration of the vehicle is detected by the acceleration sensor as a change in the movement of the vehicle during traveling. Based on the degree of follow-up of the change in the movement of the driver with respect to the acceleration of the vehicle, the determination process is executed by the physical condition determination unit. The degree of follow-up of the change in the movement of the driver with respect to the acceleration of the vehicle is different from that when the physical condition is normal at the stage where the physical condition of the driver starts to become abnormal. Therefore, according to this aspect, it is possible to detect an abnormality in the physical condition of the driver at an early stage before the abnormal state proceeds.

Another aspect of the technology disclosed herein is a driver physical condition detection method in a driver physical condition detection device that detects a physical condition of a driver driving a vehicle, and detects a change in movement of the vehicle during traveling. A vehicle detection step, a driver detection step for detecting a change in the driver's movement, a calculation step for calculating a follow-up degree of the change in the driver's movement relative to the change in the movement during the travel, and the follow-up degree. And a physical condition determination step for performing a determination process as to whether or not the driver's physical condition is abnormal, and the calculation step includes, as the follow-up degree, the driver's change with respect to a change in movement during the travel. calculating a time delay of motion changes, the physical condition determining step, wherein when the time delay is less than the reference time, the physical condition of the driver is shall be determined to be abnormal.

In this aspect, the degree of follow-up of the change in the driver's movement with respect to the change in the movement of the vehicle while traveling is calculated in the calculation step. Based on the degree of follow-up, a process for determining whether or not the driver's physical condition is abnormal is executed in the physical condition determination step. The degree of follow-up of the change in the driver's movement relative to the change in movement while the vehicle is running is different from when the physical condition is normal because the muscle strength of the neck or upper body decreases when the driver's physical condition begins to become abnormal It will be a thing. Therefore, according to this aspect, it is possible to detect an abnormality in the physical condition of the driver at an early stage before the abnormal state proceeds.
In this aspect, the time delay of the change in the driver's movement relative to the change in the movement while the vehicle is traveling is calculated as the tracking degree by the calculation step. When the time delay is equal to or shorter than the reference time, the physical condition determination step determines that the driver's physical condition is abnormal. When the muscle strength of the neck or upper body decreases at the stage where the driver's physical condition begins to become abnormal, the time delay of the change in the driver's movement with respect to the change in the movement while the vehicle is traveling is reduced. Therefore, according to this aspect, it is possible to detect an abnormality in the physical condition of the driver at an early stage before the abnormal state proceeds.

  According to one aspect of the present disclosure, since the determination process of whether or not the driver's physical condition is abnormal is executed based on the degree of tracking of the change in the driver's movement with respect to the change in the movement of the vehicle while traveling, An abnormality in the physical condition of the driver can be detected at an early stage before the abnormal state progresses.

It is a block diagram showing roughly the composition of vehicles in which the driver physical condition detection device of a 1st embodiment is carried. It is a figure which shows roughly an example of the pressure distribution of the driver | operator who seated on the driver | operator's seat detected by a seat sensor. It is a figure which shows roughly an example of the time data of the acceleration in the left-right direction of a vehicle, the time data of the driver | operator's gravity center position preserve | saved in memory by the gravity center position calculating part, and these cross correlations. It is a figure which shows roughly an example of the time data of the acceleration in the left-right direction of a vehicle, the time data of the driver | operator's gravity center position preserve | saved in memory by the gravity center position calculating part, and these cross correlations. It is a flowchart which shows roughly an example of the acceleration acquisition procedure of the vehicle in the driver | operator's physical condition detection apparatus of 1st Embodiment. It is a flowchart which shows roughly an example of the procedure which calculates the learning value of a time delay in the driver | operator's physical condition detection apparatus of 1st Embodiment. It is a flowchart which shows roughly an example of the procedure which determines a driver | operator's physical condition in the driver | operator physical condition detection apparatus of 1st Embodiment. It is a block diagram which shows roughly the structure different from FIG. 1 of the vehicle by which the driver | operator's physical condition detection apparatus of the said 1st Embodiment is mounted. It is a flowchart which shows roughly an example of the procedure which determines a driver | operator's physical condition in the structure of FIG. It is a block diagram which shows roughly the structure of the vehicle by which the driver | operator's physical condition detection apparatus of 2nd Embodiment is mounted. It is a figure which shows roughly the time change of the acceleration in the left-right direction of a motion of a vehicle, and the time change of the acceleration in the left-right direction of a driver | operator's head movement. It is a figure which shows roughly the time change of the acceleration in the left-right direction of a motion of a vehicle, and the time change of the acceleration in the left-right direction of a driver | operator's head movement. It is a figure which shows roughly an example of the time data of the acceleration in the left-right direction of a vehicle, and the time data of the acceleration in the left-right direction of a driver | operator's head. It is a figure which shows roughly an example of the time data of the acceleration in the left-right direction of a vehicle, and the time data of the acceleration in the left-right direction of a driver | operator's head. It is a flowchart which shows roughly an example of the acceleration acquisition procedure of the head in the driver | operator's physical condition detection apparatus of this 2nd Embodiment. It is a flowchart which shows roughly an example of the procedure which calculates the learning value of a time delay in the driver | operator's physical condition detection apparatus of this 2nd Embodiment. It is a flowchart which shows roughly an example of the procedure which determines a driver | operator's physical condition in the driver | operator physical condition detection apparatus of this 2nd Embodiment. It is a block diagram which shows roughly the structure different from FIG. 10 of the vehicle by which the driver | operator's physical condition detection apparatus of the said 2nd Embodiment is mounted. It is a flowchart which shows roughly an example of the procedure which determines a driver | operator's physical condition in the structure of FIG.

(Focus point of one aspect according to the present disclosure)
First, an aspect of one aspect according to the present disclosure will be described. The present inventor has found that the degree of follow-up of the driver's movement with respect to a change in the movement of the vehicle is different depending on whether the driver's physical condition is normal or abnormal while repeating various experiments.

  The inventor interprets this difference as follows. That is, a driver who is in good physical condition expects acceleration in the left-right direction from the vehicle when entering a curve from a straight road. Therefore, the muscle strength of the neck or upper body is used to resist the lateral acceleration applied from the vehicle. As a result, the degree of follow-up of the driver's movement with respect to a change in the movement of the vehicle decreases.

  On the other hand, when the driver is in poor physical condition, the muscle strength of the neck particularly decreases due to a slight decrease in consciousness, and as a result, the muscle strength of the upper body decreases. For this reason, the degree of follow-up of the driver's movement with respect to a change in the movement of the vehicle increases. Here, the “left-right direction of the vehicle” is a direction orthogonal to the front-rear direction of the vehicle within a horizontal plane. In other words, the “left-right direction of the vehicle” is a direction perpendicular to the traveling direction of the vehicle traveling on the straight road in the horizontal plane. The “left-right direction of the vehicle” can also be referred to as “vehicle width direction” or “vehicle lateral direction”.

  In the following, an experiment was conducted with a subject who was deprived of vision and hearing seated in a passenger seat to simulate a driver with poor physical condition. Since the visual and auditory senses are deprived, the subject cannot expect that the acceleration in the left-right direction is applied from the vehicle when entering the curve. For this reason, it becomes difficult to resist the acceleration in the left-right direction applied from the vehicle using the muscle strength of the neck or upper body. In this way, a driver with poor physical condition in which the muscle strength of the neck or upper body is reduced is realized in a simulated manner.

  From the above consideration, the present inventor has found that it is possible to detect a driver's poor physical condition at an early stage by examining the degree of tracking of the driver's movement with respect to a change in the movement of the vehicle.

(Embodiment)
Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In each figure, the same numerals are given to the same element, and explanation is omitted suitably.

(First embodiment)
FIG. 1 is a block diagram schematically showing the configuration of a vehicle on which the driver physical condition detection device of the first embodiment is mounted. The vehicle 10 is a four-wheeled vehicle, for example. As shown in FIG. 1, the vehicle 10 includes an acceleration sensor 102, a seat sensor 103, an alarm sound generator 201, an alarm lamp 202, and an electronic control unit (ECU) 300.

  The acceleration sensor 102 (an example of a vehicle detection unit) detects the acceleration of the vehicle 10 in, for example, three orthogonal axes. The acceleration sensor 102 outputs the detected acceleration of the vehicle 10 to the ECU 300. The seat sensor 103 is disposed on the seat surface portion of the driver's seat. The seat sensor 103 includes, for example, 32 × 32 piezoelectric elements, and detects the pressure distribution of the driver seated on the driver's seat. The sheet sensor 103 outputs detection data to the ECU 300.

  The alarm sound generator 201 includes, for example, an electronic buzzer and generates an alarm sound for the driver. The alarm lamp 202 includes, for example, a light emitting diode, and displays an alarm to the driver. The alarm lamp 202 is not limited to a dedicated lamp, and may be used as an alarm lamp by blinking a meter on an instrument panel.

  ECU 300 controls the overall operation of vehicle 10. ECU 300 includes a memory 310, a central processing unit (CPU) 320, and other peripheral circuits. The memory 310 is composed of, for example, a semiconductor memory such as a flash memory, a hard disk, or another storage element. The memory 310 includes a memory that stores a program, a memory that temporarily stores data, and the like. The memory 310 may be composed of a single memory having an area for storing a program, an area for temporarily storing data, and the like.

  The CPU 320 operates in accordance with a program stored in the memory 310, whereby an acceleration control unit 321, a physical condition determination unit 322, an alarm control unit 325, a pressure data acquisition unit 351, a centroid position calculation unit 352, a cross correlation calculation unit 361, And the learning value control unit 362.

  The acceleration control unit 321 acquires acceleration in the left-right direction of the movement of the vehicle 10 from, for example, acceleration data in the orthogonal three-axis directions of the vehicle 10 output from the acceleration sensor 102 at predetermined time (for example, 100 msec). The acceleration control unit 321 stores time data of acceleration in the left-right direction of the movement of the vehicle 10 for a predetermined time (in this embodiment, for example, 10 seconds). For example, if the predetermined time is 10 seconds and acceleration data is acquired from the acceleration sensor 102 every 100 msec, 100 acceleration time data are stored in the memory 310.

  The pressure data acquisition unit 351 acquires pressure data of each piezoelectric element output from the sheet sensor 103. The pressure data acquisition unit 351 outputs the acquired pressure data to the gravity center position calculation unit 352. The center-of-gravity position calculation unit 352 calculates the center-of-gravity position of the driver in the left-right direction of the vehicle 10 using the pressure data from the pressure data acquisition unit 351. The center-of-gravity position calculation unit 352 stores data on the center-of-gravity position of the driver for a predetermined time in the memory 310. In this embodiment, the predetermined time is, for example, 10 seconds.

  FIG. 2 is a diagram schematically illustrating an example of the pressure distribution of the driver seated on the driver's seat detected by the seat sensor 103. FIG. 2 shows isobaric lines having the same pressure value. In FIG. 2, the pressure value at the center position of the dense isobaric line is the highest.

  The sheet sensor 103 outputs a set of coordinates and pressure values of each piezoelectric element to the ECU 300 as serial data, for example. The pressure data acquisition unit 351 processes the received serial data and outputs a set of coordinates and pressure values of each piezoelectric element to the barycentric position calculation unit 352. The center-of-gravity position calculation unit 352 calculates the center-of-gravity position of the driver in the left-right direction of the vehicle 10 (the X-axis direction in FIG. 2) from the set of coordinates and pressure values of each piezoelectric element. The center-of-gravity position calculation unit 352 calculates the center-of-gravity position of the driver, for example, as the distance from the origin on the X axis.

  Returning to FIG. 1, the cross-correlation calculation unit 361 includes time data of acceleration in the left-right direction of the vehicle 10 stored in the memory 310 by the acceleration control unit 321 and the driver stored in the memory 310 by the gravity center position calculation unit 352. The cross-correlation with the time data of the centroid position is calculated. The cross-correlation is one of the convolution formulas that convolve two functions (in this embodiment, a function representing time data of acceleration in the left-right direction of the vehicle 10 and a function representing time data of the center of gravity of the driver). Is obtained by conversing the signal array in the reverse order. The cross correlation calculation unit 361 calculates a time delay from the obtained cross correlation.

  FIGS. 3 and 4 schematically show examples of acceleration time data in the left-right direction of the vehicle 10, time data of the driver's center of gravity position stored in the memory 310 by the center-of-gravity position calculation unit 352, and cross-correlation examples thereof. FIG. The section (A) in FIGS. 3 and 4 shows time data of the acceleration CA in the left-right direction of the vehicle 10 and time data of the center-of-gravity position GC in the left-right direction of the driver. 3 and 4, the horizontal axis represents time [seconds], the left vertical axis represents acceleration [G], and the right vertical axis represents the gravity center position [cm]. The section (A) in FIG. 3 shows the case of a driver who is in a normal physical condition, and the section (A) in FIG. 4 shows the case of a driver in an abnormal physical condition.

  The section (B) in FIGS. 3 and 4 shows the cross-correlation MC between the time data of the acceleration CA shown in the section (A) and the time data of the barycentric position GC. 3 and 4, the horizontal axis represents time delay, and the vertical axis represents the magnitude of the correlation value.

  Section (B) in FIG. 3 shows that there is a negative correlation because the peak value of the correlation value is close to (−1), and the peak value of the correlation value is shifted with respect to the time delay “0”. It represents that there is a time delay TD. Section (B) in FIG. 4 indicates that there is a negative correlation because the peak value of the correlation value is close to (−1), and the time delay “0” matches the peak value of the correlation value. Indicates no time delay.

  As shown in the section (A) of FIGS. 3 and 4, when the vehicle 10 traveling on the straight road enters the curve, the acceleration in the left-right direction of the vehicle 10 increases, and the acceleration increases accordingly. Thus, the position of the center of gravity in the left-right direction of the driver moves. When the acceleration in the left-right direction of the vehicle 10 increases, the driver who is in a normal physical condition tries to maintain a good field of view by putting power on the upper body so that the field of view is not hindered by the increase in the acceleration. For this reason, as shown in section (B) of FIG. 3, a time delay TD occurs in the cross-correlation MC between the time data of the acceleration CA and the time data of the center of gravity position GC.

  On the other hand, when the acceleration in the left-right direction of the vehicle 10 increases, it is difficult for a driver with poor physical condition to apply force to the upper body, so that the position of the center of gravity moves according to the increase in acceleration. As a result, as shown in section (B) of FIG. 4, there is almost no time delay in the time data of the center of gravity position GC with respect to the time data of the acceleration CA.

  Returning to FIG. 1, the learning value control unit 362 considers that the physical condition of the driver is normal until a predetermined time elapses after the ignition switch of the vehicle 10 is turned on. The average value of the time delay of the cross-correlation between the time data of the acceleration CA and the time data of the center of gravity position GC obtained in the above is stored in the memory 310 as a learning value.

  The physical condition determination unit 322 compares the time delay of the cross-correlation between the time data of the acceleration CA in the left-right direction of the vehicle 10 and the time data of the center of gravity position GC of the driver with the learning value stored in the memory 310, Based on the comparison result, it is determined whether or not the driver's physical condition is abnormal. Specifically, the physical condition determination unit 322 determines that the physical condition of the driver is abnormal if the time delay of the cross-correlation between the time data of the acceleration CA and the time data of the gravity center position GC is equal to or less than K1 times the learning value. judge. The coefficient K1 is a value less than 1, and in this embodiment, for example, K1 = 0.5 is set. If the physical condition determination unit 322 determines that the driver's physical condition is abnormal, the physical condition determination unit 322 notifies the alarm control unit 325 that the physical condition of the driver is abnormal.

  The physical condition determination unit 322 determines whether or not the driver's physical condition is abnormal only when the acceleration in the left-right direction of the movement of the vehicle 10 is equal to or greater than a predetermined acceleration threshold ACth. The reason for this is that when the acceleration in the left-right direction of the movement of the vehicle 10 is small, the correlation between the time data of the acceleration CA and the time data of the center of gravity position GC depends on whether the driver is normal or abnormal. This is because there is no significant difference in time delay. In this embodiment, for example, ACth = 0.1 [G] is set.

  When notified from the physical condition determination unit 322 that the physical condition of the driver is abnormal, the alarm control unit 325 activates the alarm sound generator 201 and blinks the alarm lamp 202 to alert the driver. . For example, the alarm control unit 325 may assist the driver's driving by operating a brake to decelerate or stop the vehicle 10 or control the steering wheel to move the vehicle 10 to the road shoulder.

  FIG. 5 is a flowchart schematically illustrating an example of a vehicle acceleration acquisition procedure in the driver physical condition detection device of the first embodiment. The flow in FIG. 5 is executed every predetermined time (for example, 100 msec). In step S <b> 2500, the acceleration control unit 321 acquires the acceleration in the left-right direction of the movement of the vehicle 10 from the acceleration data of the vehicle 10 output from the acceleration sensor 102 in, for example, orthogonal three-axis directions. In step S2510, the acceleration control unit 321 stores time data of acceleration in the left-right direction of the movement of the vehicle 10 for a predetermined time (in this embodiment, for example, 10 seconds). That is, when new acceleration data is obtained, the acceleration control unit 321 deletes the oldest acceleration data from the memory 310 so that the acceleration data for a predetermined time is stored in the memory 310. Yes. Thereafter, the process of FIG. 5 ends.

  FIG. 6 is a flowchart schematically showing an example of a procedure for calculating a learning value of time delay in the driver physical condition detection device of the first embodiment. The flow in FIG. 6 is executed every predetermined time (for example, 100 msec).

  In step S <b> 2600, the pressure data acquisition unit 351 acquires pressure data from the sheet sensor 103. In step S2610, the center-of-gravity position calculation unit 352 calculates the center-of-gravity position of the driver in the left-right direction. The center-of-gravity position calculation unit 352 stores time data of the driver's center-of-gravity position for a predetermined time in the memory 310.

  In step S2620, learned value control unit 362 determines whether or not a predetermined time has elapsed since the ignition switch of vehicle 10 was turned on.

  If the predetermined time has not elapsed since the ignition switch of vehicle 10 was turned on (NO in step S2620), the process proceeds to S2630. On the other hand, if a predetermined time has elapsed since the ignition switch of vehicle 10 was turned on (YES in step S1720), the process in FIG. 6 ends. That is, if NO in step S2620, it is considered that the driver's physical condition is normal, the process proceeds to step S2630, and a process for obtaining a learning value is performed. On the other hand, if “YES” in the step S2620, the process of FIG. 6 ends without performing the process of obtaining the learning value.

  In step S2630, the cross-correlation calculation unit 361 calculates the cross-correlation from the time data of the center of gravity position stored in the memory 310 and the time data of the vehicle acceleration, and calculates the time delay of the peak value of the correlation value. . In step S2640, the learning value control unit 362 stores the calculated average value of time delay in the memory 310 as a learning value. Thereafter, the process of FIG. 6 ends.

  FIG. 7 is a flowchart schematically showing an example of a procedure for determining the physical condition of the driver in the driver physical condition detection device of the first embodiment. The flow in FIG. 7 is executed every predetermined time (for example, 100 msec).

  In step S <b> 2700, the physical condition determination unit 322 obtains the latest acceleration CAN from the memory 310 among the time data of acceleration in the left-right direction of the vehicle 10 stored in the memory 310 (in this embodiment, for example, data for 10 seconds). get. In step S2710, the physical condition determination unit 322 determines whether the acquired absolute value of the latest acceleration CAn in the left-right direction of the vehicle is greater than or equal to the acceleration threshold ACth. If the absolute value of the latest acceleration CAn in the left-right direction of the vehicle is less than the acceleration threshold ACth (NO in step S2710), the process in FIG. 7 ends. On the other hand, if the absolute value of the latest acceleration CAn in the left-right direction of the vehicle is greater than or equal to acceleration threshold ACth (YES in step S2710), the process proceeds to step S2720.

  As described above, in this embodiment, for example, ACth = 0.1 [G]. For example, in section (A) of FIG. 3, time t1 is the time when the absolute value of acceleration CA in the left-right direction of the vehicle is 0.1 [G] or more. Thereafter, time t2 is the time when the absolute value of acceleration CA in the left-right direction of the vehicle becomes less than 0.1 [G]. The time t3 is a time when the absolute value of the acceleration CA in the left-right direction of the vehicle becomes 0.1 [G] or more thereafter. Thereafter, time t4 is the time when the absolute value of the acceleration CA in the left-right direction of the vehicle becomes less than 0.1 [G].

  Therefore, for example, in the section (A) of FIG. 3, during the period from the beginning to the time t1, NO is determined in step S2710, and the process of FIG. 7 ends. Thereafter, between time t1 and time t2 and between time t3 and time t4, YES is obtained in step S2710, and the process proceeds to step S2720.

  For example, in section (A) of FIG. 4, time t11 is the time when the absolute value of acceleration CA in the left-right direction of the vehicle becomes 0.1 [G] or more. Thereafter, time t12 is the time when the absolute value of acceleration CA in the left-right direction of the vehicle becomes less than 0.1 [G]. Thereafter, time t13 is the time when the absolute value of acceleration CA in the left-right direction of the vehicle becomes equal to or greater than 0.1 [G]. Thereafter, time t14 is a time when the absolute value of acceleration CA in the left-right direction of the vehicle becomes less than 0.1 [G].

  Therefore, for example, in the section (A) of FIG. 4, NO is determined in step S2710 from the beginning to the time t11, and the process of FIG. 7 ends. Thereafter, between time t11 and time t12 and between time t13 and time t14, YES is obtained in step S2710, and the process proceeds to step S2720.

  Returning to FIG. 7, in step S2720, the cross-correlation calculating unit 361 calculates the cross-correlation from the time data of the center of gravity and the time data of the vehicle acceleration, and calculates the time delay of the peak value of the correlation value. In step S <b> 2730, the physical condition determination unit 322 acquires a time-lag learning value stored in the memory 310 from the memory 310.

  In step S2740, the physical condition determination unit 322 determines whether the time delay calculated in step S2720 is equal to or less than K1 times the learning value acquired in step S2730. If the calculated time delay is greater than K1 times the learned value (NO in step S2740), the process in FIG. 7 ends. If the calculated time delay is equal to or less than K1 times the learned value (YES in step S2740), the process proceeds to step S2750.

  In step S2750, the physical condition determination unit 322 determines that the physical condition of the driver is abnormal, and notifies the alarm control unit 325 accordingly. In step S2760, the alarm control unit 325 operates the alarm sound generator 201 and the alarm lamp 202 to notify the driver that the physical condition is abnormal, and the process in FIG. 7 ends.

  As described above, in the first embodiment, the cross-correlation calculating unit 361 calculates the cross-correlation from the time data of the gravity center position and the time data of the vehicle acceleration, and calculates the time delay of the peak value of the correlation value. . The physical condition determination unit 322 determines that the physical condition of the driver is abnormal if the time delay is equal to or less than K1 times the learning value. When the driver is in poor physical condition, the muscle strength of the neck or upper body is reduced due to a slight decrease in consciousness, and it is difficult to apply force to the upper body. Therefore, when the acceleration in the left-right direction of the vehicle 10 increases, Will follow and move. As a result, there is almost no time delay in the time data of the center of gravity position GC of the driver with respect to the time data of the acceleration CA of the vehicle. Therefore, according to the first embodiment, the abnormality of the driver's physical condition can be detected at an early stage before the abnormal state of the driver's physical condition progresses.

  In the first embodiment, the learning value control unit 362 considers that the driver's physical condition is normal until a predetermined time elapses after the ignition switch of the vehicle 10 is turned on. Thus, the average value of the time delay obtained during this time is stored in the memory 310 as a learning value. Thus, since the learning value is obtained each time the ignition switch of the vehicle 10 is turned on, an appropriate learning value can be obtained for the driver. Therefore, according to this 1st Embodiment, the quality of a driver | operator's physical condition can be determined correctly.

  In the first embodiment, the time delay of the cross correlation is compared with the learning value, but the present invention is not limited to this. For example, the time delay of the cross correlation may be compared with a predetermined determination threshold.

  FIG. 8 is a block diagram schematically showing a configuration different from FIG. 1 of the vehicle on which the driver physical condition detection device of the first embodiment is mounted. The CPU 320 in FIG. 8 does not include the learning value control unit 362 provided in the CPU 320 in FIG. The physical condition determination unit 322 of FIG. 8 determines that the physical condition of the driver is abnormal if the time delay of the cross-correlation is equal to or less than the determination threshold value stored in the memory 310 in advance. The determination threshold value is predetermined experimentally, for example, and stored in the memory 310 in advance. The determination threshold may be about 3 seconds, for example.

  FIG. 9 is a flowchart schematically showing an example of a procedure for determining the physical condition of the driver in the configuration of FIG. The flow in FIG. 9 is executed every predetermined time (for example, 100 msec).

  Steps S2700, S2710, and S2720 in FIG. 9 are the same as steps S2700, S2710, and S2720 in FIG. In step S <b> 2900 subsequent to step S <b> 2720, the physical condition determination unit 322 acquires a time delay determination threshold value from the memory 310.

  In step S2910, the physical condition determination unit 322 determines whether or not the time delay calculated in step S2720 is equal to or less than a determination threshold value. If the calculated time delay is greater than the determination threshold (NO in step S2910), the process in FIG. 9 ends. If the calculated time delay is equal to or smaller than the determination threshold value (YES in step S2910), the process proceeds to step S2750. Steps S2750 and S2760 are the same as steps S2750 and S2760 in FIG. 7, respectively.

  Thus, even when the time delay of the cross-correlation is compared with a predetermined determination threshold, as in the first embodiment, at the early stage before the abnormal state of the driver's physical condition progresses, Abnormal physical condition can be detected.

(Second Embodiment)
FIG. 10 is a block diagram schematically showing the configuration of a vehicle on which the driver physical condition detection device of the second embodiment is mounted. In the first embodiment, the driver's physical condition is determined using the time data of the driver's center of gravity, whereas in the second embodiment, the driver's head acceleration time data is used. To determine the physical condition of the driver.

  The vehicle 10 of the second embodiment includes a camera 101, an acceleration sensor 102, an alarm sound generator 201, an alarm lamp 202, and an ECU 300.

  The camera 101 (an example of a driver detection unit) is attached to, for example, a ceiling in front of a driver's seat in the vehicle 10 so that the optical axis of the camera 101 faces the driver's seat of the vehicle 10. The camera 101 images the driver of the vehicle 10 from the front and images the head of the driver moving in the left-right direction. The camera 101 outputs the captured frame image to the ECU 300, for example, every 1/60 seconds. Alternatively, the camera 101 may be attached to the ceiling above the driver's seat in the vehicle 10 so that the optical axis of the camera 101 faces the driver's seat of the vehicle 10. Further alternatively, a plurality of cameras may be attached to the ceiling or the like of the interior of the vehicle 10 so that each optical axis faces the driver's seat of the vehicle 10. The camera 101 only needs to be attached to the interior of the vehicle 10 so that the movement of the driver's head in the left-right direction can be imaged.

  In the second embodiment, the CPU 320 operates according to a program stored in the memory 310, thereby causing an acceleration control unit 321, a physical condition determination unit 322, an acceleration calculation unit 323, a head detection unit 324, an alarm control unit 325, and a mutual control. It functions as a correlation calculation unit 361 and a learning value control unit 362.

  The head detection unit 324 detects the driver's head from the frame image captured by the camera 101, for example, by template matching. The head detection unit 324 stores, for example, the position coordinates of the center of the driver's head in the imaging range of the camera 101 in the memory 310 for each frame image. The head detection unit 324 stores time data of the position coordinates of the driver's head for a predetermined time in the memory 310. For example, when a predetermined time is 1 second and a frame image is output from the camera 101 every 1/60 seconds, time data of 60 driver's head position coordinates is stored in the memory 310. The Rukoto.

  The acceleration calculation unit 323 uses the time data of the position coordinates of the driver's head stored in the memory 310 to calculate the acceleration in the left-right direction of the movement of the driver's head. For example, the acceleration calculation unit 323 calculates the movement distance between the frame images from the position coordinates of the head for each frame image, and calculates the acceleration from the amount of change of the calculated movement distance for each frame image. The acceleration calculation unit 323 stores time data of acceleration of the movement of the driver's head for a predetermined time in the memory 310. As described above, when a predetermined time is 1 second, for example, and a frame image is output from the camera 101 every 1/60 seconds, 60 driver's head movement acceleration time data Is stored in the memory 310.

  The physical condition determination unit 322 determines whether or not there is a time delay in the change in acceleration in the horizontal direction of the movement of the head of the driver with respect to the change in acceleration in the horizontal direction of the movement of the vehicle 10. It is determined whether or not the physical condition is abnormal.

  FIGS. 11 and 12 are diagrams schematically showing a temporal change in the acceleration CA in the left-right direction of the movement of the vehicle 10 and a temporal change in the acceleration HD in the left-right direction of the movement of the driver's head. FIG. 11 shows a case where the driver's physical condition is normal, and FIG. 12 shows a case where the driver is in poor physical condition.

  When the time delay T1 of the change in the head acceleration HD with respect to the change in the vehicle acceleration CA in FIG. 11 is compared with the time delay T2 in the change in the head acceleration HD with respect to the change in the vehicle acceleration CA in FIG. > T2.

  The reason is considered as follows. That is, when the driver is in poor physical condition, the neck muscle strength is reduced, and the head acceleration HD changes following the change in the acceleration CA of the vehicle. Therefore, the time delay T2 is relatively small. On the other hand, when the driver's physical condition is normal, the time delay T1 of the change in the head acceleration HD is relatively large because the driver tries to resist the change in the acceleration CA of the vehicle by putting effort into the muscle strength of the neck. large.

  Returning to FIG. 10, the physical condition determination unit 322 determines whether the physical condition of the driver is abnormal only when the acceleration in the left-right direction of the movement of the vehicle 10 is equal to or greater than a predetermined acceleration threshold ACth. . The reason for this is that when the acceleration of the movement of the vehicle 10 in the left-right direction is small, the time delay of the change in the head acceleration HD with respect to the change in the vehicle acceleration CA, depending on whether the driver is normal or abnormal. This is because no significant difference occurs. In this embodiment, for example, ACth = 0.1 [G] is set.

  The cross-correlation calculation unit 361 includes acceleration time data in the left-right direction of the vehicle 10 stored in the memory 310 by the acceleration control unit 321 and left-right direction of the driver's head stored in the memory 310 by the acceleration calculation unit 323. The cross-correlation with acceleration time data is calculated. As described above, the cross-correlation has two functions (in this embodiment, a function representing time data of acceleration in the left-right direction of the vehicle 10 and a function representing time data of acceleration in the left-right direction of the driver's head). Of the convolution formulas, the signal sequence of one of the functions is reversed and the convolution is obtained. The cross correlation calculation unit 361 calculates a time delay from the obtained cross correlation.

  13 and 14 are diagrams schematically illustrating an example of time data of acceleration CA in the left-right direction of the vehicle 10 and time data of acceleration HD in the left-right direction of the driver's head. 13 and 14, the horizontal axis represents time [seconds], the left vertical axis represents head acceleration [G], and the right vertical axis represents vehicle 10 acceleration [G]. FIG. 13 shows the case of a driver with normal physical condition, and FIG. 14 shows the case of a driver with abnormal physical condition.

  As shown in FIG. 13 and FIG. 14, when the vehicle 10 traveling on the straight road enters the curve, the acceleration in the left-right direction of the vehicle 10 increases, and according to the increase in the acceleration, the driver's The acceleration in the left-right direction of the head increases. A driver who is in good physical condition tries to prevent the head from shaking by applying force to the neck muscles so that the head does not shake due to the increase in acceleration when the acceleration in the left-right direction of the vehicle 10 increases. To do. For this reason, a time delay occurs in the cross-correlation between the time data of the acceleration CA of the vehicle and the time data of the acceleration HD of the driver's head.

  On the other hand, when the acceleration in the left-right direction of the vehicle 10 increases, the driver with poor physical condition decreases the neck muscle strength, so that the acceleration of the head follows the increase in the acceleration of the vehicle. As a result, there is almost no time delay in the time data of the head acceleration HD with respect to the time data of the acceleration CA of the vehicle.

  Returning to FIG. 10, the learning value control unit 362 assumes that the physical condition of the driver is normal until a predetermined time elapses after the ignition switch of the vehicle 10 is turned on. The average value of the time delay of the cross-correlation between the time data of the acceleration CA of the vehicle 10 and the time data of the head acceleration HD obtained in the above is stored in the memory 310 as a learning value.

  The physical condition determination unit 322 includes a time delay of cross-correlation between time data of acceleration CA in the left-right direction of the vehicle 10 and time data of acceleration HD in the left-right direction of the driver's head, and a learning value stored in the memory 310. And based on the comparison result, it is determined whether or not the driver's physical condition is abnormal. Specifically, the physical condition determination unit 322 determines that the driver has a time delay of cross-correlation between time data of the vehicle acceleration CA and time data of the driver's head acceleration HD less than or equal to K2 times the learned value. Is determined to be abnormal. The coefficient K2 is a value less than 1, and in this embodiment, for example, K2 = 0.5. The coefficient K2 may be the same value as the coefficient K1 or a different value. If the physical condition determination unit 322 determines that the driver's physical condition is abnormal, the physical condition determination unit 322 notifies the alarm control unit 325 that the physical condition of the driver is abnormal.

  The physical condition determination unit 322 determines whether or not the driver's physical condition is abnormal only when the acceleration in the left-right direction of the movement of the vehicle 10 is equal to or greater than a predetermined acceleration threshold ACth. This is because when the acceleration of the movement of the vehicle 10 in the left-right direction is small, the time data of the acceleration CA and the time data of the head acceleration HD are different depending on whether the driver is normal or abnormal. This is because there is no significant difference in the correlation time delay. In this embodiment, for example, ACth = 0.1 [G] is set.

  The vehicle acceleration acquisition procedure by the acceleration control unit 321 in the second embodiment is the same as the procedure in the first embodiment shown in FIG.

  FIG. 15 is a flowchart schematically showing an example of a head acceleration acquisition procedure in the driver physical condition detection device of the second embodiment. The flow in FIG. 15 is executed at predetermined time intervals (for example, every frame image output from the camera 101, that is, every 1/60 sec in this embodiment).

  In step S <b> 3300, the head detection unit 324 acquires frame image data captured by the camera 101. In step S3310, the head detection unit 324 detects the driver's head from the acquired frame image, and calculates, for example, the position coordinates of the center of the driver's head. In step S3320, the head detecting unit 324 stores time data of the position coordinates of the driver's head for a predetermined time in the memory 310 for each frame image. That is, when the new position coordinate data is obtained, the head detection unit 324 deletes the oldest position coordinate data from the memory 310 and stores the position coordinate data for a predetermined time in the memory 310. I try to do it.

  In step S3330, the acceleration calculation unit 323 calculates the acceleration in the left-right direction of the movement of the driver's head using the time data of the position coordinates of the driver's head stored in the memory 310. In step S <b> 3340, the acceleration calculation unit 323 stores the time data of the acceleration of the driver's head movement for a predetermined time in the memory 310. Similar to step S3320, when new acceleration data is obtained, the acceleration calculation unit 323 deletes the oldest acceleration data from the memory 310, and the acceleration data for a predetermined time is stored in the memory 310. I try to do it. Thereafter, the process of FIG. 15 ends.

  FIG. 16 is a flowchart schematically showing an example of a procedure for calculating a learning value of time delay in the driver physical condition detection device of the second embodiment. The flow in FIG. 16 is executed every predetermined time (for example, 100 msec).

  In step S3400, learned value control unit 362 determines whether or not a predetermined time has elapsed since the ignition switch of vehicle 10 was turned on.

  If the predetermined time has not elapsed since the ignition switch of the vehicle 10 was turned on (NO in step S3400), the process proceeds to S3410. On the other hand, if a predetermined time has elapsed since the ignition switch of vehicle 10 was turned on (YES in step S3400), the process in FIG. 16 ends. That is, if NO in step S3400, the driver's physical condition is considered normal, and the process proceeds to step S3410 to perform a process for obtaining a learning value. On the other hand, if “YES” in the step S3400, the process of FIG. 16 ends without performing the process of obtaining the learning value.

  In step S3410, the cross-correlation calculating unit 361 calculates the cross-correlation from the time data of the head acceleration stored in the memory 310 and the time data of the vehicle acceleration, and the time delay of the peak value of the correlation value. Is calculated. In step S3420, the learning value control unit 362 stores the calculated average value of time delay in the memory 310 as a learning value. Thereafter, the process of FIG. 16 ends.

  FIG. 17 is a flowchart schematically showing an example of a procedure for determining the physical condition of the driver in the driver physical condition detection device of the second embodiment. The flow in FIG. 17 is executed every predetermined time (for example, 100 msec).

  Steps S2700 and S2710 are the same as steps S2700 and S2710 in FIG. In step S3500 following step S2710, the cross-correlation calculation unit 361 calculates a cross-correlation from the head acceleration time data and the vehicle acceleration time data, and calculates a time delay of the peak value of the correlation value. In step S <b> 3510, the physical condition determination unit 322 acquires a time-lag learning value stored in the memory 310 from the memory 310.

  In step S3520, the physical condition determination unit 322 determines whether the time delay calculated in step S3500 is equal to or less than K2 times the learning value acquired in step S3510. If the calculated time delay is larger than K2 times the learned value (NO in step S3520), the process in FIG. 17 ends. If the calculated time delay is equal to or less than K2 times the learning value (YES in step S3520), the process proceeds to step S2750. Steps S2750 and S2760 are the same as steps S2750 and S2760 in FIG.

  As described above, in the second embodiment, the cross-correlation calculating unit 361 calculates the cross-correlation from the time data of the head acceleration and the time data of the vehicle acceleration, and calculates the time delay of the peak value of the correlation value. calculate. The physical condition determination unit 322 determines that the physical condition of the driver is abnormal if the time delay is equal to or less than K2 times the learned value. When the driver is in a poor physical condition, the muscle strength of the neck or upper body is reduced due to a slight decrease in consciousness, and it is difficult to apply force to the upper body. It will follow. As a result, there is almost no time delay in the time data of the acceleration HD of the driver's head with respect to the time data of the acceleration CA of the vehicle. Therefore, according to the second embodiment, the abnormality of the driver's physical condition can be detected at an early stage before the abnormal state of the driver's physical condition progresses.

  Further, in the second embodiment, as in the first embodiment, the learned value control unit 362 allows the driver to wait until a predetermined time elapses after the ignition switch of the vehicle 10 is turned on. The average value of the time delay obtained during this time is stored in the memory 310 as a learning value. Thus, since the learning value is obtained each time the ignition switch of the vehicle 10 is turned on, an appropriate learning value can be obtained for the driver. Therefore, according to the second embodiment, the quality of the driver's physical condition can be accurately determined as in the first embodiment.

  In the second embodiment, the time delay of the cross-correlation is compared with the learning value, but is not limited to this. For example, the time delay of the cross correlation may be compared with a predetermined determination threshold.

  FIG. 18 is a block diagram schematically showing a configuration different from FIG. 10 of the vehicle on which the driver physical condition detection device of the second embodiment is mounted. The CPU 320 in FIG. 18 does not include the learning value control unit 362 provided in the CPU 320 in FIG. The physical condition determination unit 322 in FIG. 18 determines that the physical condition of the driver is abnormal if the time delay of the cross-correlation is equal to or less than a determination threshold value stored in the memory 310 in advance. The determination threshold value is predetermined experimentally, for example, and stored in the memory 310 in advance. The determination threshold may be about 3 seconds, for example.

  FIG. 19 is a flowchart schematically showing an example of a procedure for determining the physical condition of the driver in the configuration of FIG. The flow in FIG. 19 is executed every predetermined time (for example, 100 msec).

  Steps S2700 and S2710 in FIG. 19 are the same as steps S2700 and S2710 in FIG. 7, respectively. Step S3500 in FIG. 19 is the same as step S3500 in FIG. Steps S2900 and S2910 in FIG. 19 are the same as steps S2900 and S2910 in FIG. 9, respectively. Steps S2750 and S2760 in FIG. 19 are the same as steps S2750 and S2760 in FIG. 7, respectively.

  Thus, even if the time delay of the cross-correlation is compared with a predetermined determination threshold, as in the second embodiment, at the early stage before the abnormal state of the driver's physical condition progresses, Abnormal physical condition can be detected.

DESCRIPTION OF SYMBOLS 101 Camera 102 Acceleration sensor 103 Sheet sensor 310 Memory 321 Acceleration control part 322 Physical condition determination part 323 Acceleration calculation part 324 Head detection part 351 Pressure data acquisition part 352 Center of gravity position calculation part 361 Cross correlation calculation part 362 Learning value control part

Claims (8)

A driver physical condition detection device for detecting a physical condition of a driver driving a vehicle,
A vehicle detection unit for detecting a change in movement of the vehicle during travel;
A driver detector for detecting a change in the movement of the driver;
A calculation unit that calculates a tracking degree of the change in the driver's movement with respect to the change in movement during the driving;
A physical condition determination unit that executes a determination process as to whether or not the driver's physical condition is abnormal based on the following degree;
Equipped with a,
The calculation unit calculates a time delay of the change of the driver's movement with respect to the change of the movement during the travel as the tracking degree,
The physical condition determination unit determines that the physical condition of the driver is abnormal when the time delay is equal to or less than a reference time;
OPERATION's physical condition sensing device. The vehicle detection unit detects a change in movement during the traveling in the left-right direction of the vehicle,
The driver detection unit detects a change in the driver's movement in the left-right direction,
The calculation unit calculates a follow-up degree of a change in the driver's movement in the left-right direction with respect to a change in the movement in the left-right direction.
The driver's physical condition detection device according to claim 1. The driver detection unit detects a change in the center of gravity position of the driver as a change in the driver's movement,
The calculation unit calculates a follow-up degree of the change in the center of gravity position with respect to the change in movement during the traveling.
The driver physical condition detection device according to claim 1 or 2. The driver detection unit detects acceleration of the driver's head as a change in the driver's movement,
The calculation unit calculates a degree of follow-up of acceleration of the head with respect to a change in movement during the running.
The driver physical condition detection device according to claim 1 or 2. The calculation unit calculates the time delay by calculating a cross-correlation between the time change of the movement during the travel and the time change of the driver's movement,
The driver's physical condition detection device according to claim 1 . The physical condition determination unit executes the determination process only when the magnitude of the change in movement during running is equal to or greater than a predetermined threshold.
The driver's physical condition detection device according to any one of claims 1 to 5 . The vehicle detection unit includes an acceleration sensor that detects an acceleration of the vehicle as a change in movement during the traveling.
The driver's physical condition detection device according to any one of claims 1 to 6 . A driver physical condition detection method in a driver physical condition detection device for detecting a physical condition of a driver driving a vehicle,
A vehicle detection step of detecting a change in movement of the vehicle during travel;
A driver detection step of detecting a change in the movement of the driver;
A calculation step of calculating a follow-up degree of the change in the driver's movement with respect to the change in movement during the driving;
Based on the degree of follow-up, a physical condition determination step for executing a determination process of whether or not the driver's physical condition is abnormal,
Equipped with a,
The calculation step calculates, as the tracking degree, a time delay of the change in the driver's movement with respect to the change in movement during the running,
The physical condition determination step determines that the physical condition of the driver is abnormal when the time delay is equal to or less than a reference time.
OPERATION's physical condition detection method.

Images